On the surface, the least interesting page on Yahoo (YHOO) is its list of press releases. But take a look sometime: Yahoo announces the formation of two to three new alliances every week. Some are for content and some for technology, and they vary in commitment and intimacy from simple alliances to outright acquisitions.

Two to three a week. According to Elizabeth Collet, Yahoo's director of business development, the press releases represent only a small fraction of the actual number of deals a company makes. With that many deals inked each week, the most important thing is nimbleness - Yahoo's partnerships take an average negotiation time of "five hours, 10 at the outset," Collet says.

Over the summer, an intern from Harvard Business School negotiated 30 agreements with digital greeting-card companies, and Yahoo Greetings was born. When the company decided it needed to add a calendar feature, it simply acquired a company whose product it liked, closed the deal on a Friday, had the (new) employees modify their code over the weekend, and launched the service at the same time it announced the acquisition - on Monday.

If you're a traditional offline business, these numbers should scare the daylights out of you. It probably takes you five to 10 hours just to get someone in your corporate legal department to call you back. But for startups and brick-and-mortar companies alike, the ability to quickly identify, form, operate and conclude relationships with other companies is critical for a successful e-business strategy.

Given the pace with which long-standing relationships throughout most industries are breaking down, every company at a minimum needs to rethink what its real partners are (and what its competitors are) and get ready to rearrange the chairs. As they do, they'll soon discover that the new partnerships are very different from the ones they're used to. There are more of them, for one thing, maybe a lot more. The relationships will also be shorter-lived. And there will be a much stronger emphasis on technology partnerships than ever before.

That last part will require significant change to your corporate culture, particularly the culture related to your IT department. The operating model for the last two decades has been vendor and customer; in the days of proprietary operating systems and applications, it was more like vendor and captives. But as IT moves increasingly from a back-office to a front-office strategic role, the nature of those relationships must also change. If you are relying on technologies, especially new and unproven ones, to launch your killer app, you must create true partnerships with your technology providers, whether that means hardware, software, communications, services or all of the above.

Some very old-fashioned companies I know of are taking this principle even further and treating their partners as joint venturers. As they create organizational units to develop and launch their e-business initiatives, many are offering equity stakes to their content, channel and technology partners. Shared upside, shared risk - that's the best way to create the right incentives for a risky new venture.

Eastman Chemical (EMN) has gone even further. It recently opened an office in Northern California (its headquarters is in Kingsport, Tenn.) to identify early-stage investments, and in a few short months has already announced deals with key business-to-business (ChemConnect) and infrastructure (WebMethods) startups.

Since e-business ventures are inherently risky, your partnerships are a hedge against great uncertainty. Just as a financial planner builds a balanced portfolio of holdings and manages to the overall performance of the portfolio, you must build a portfolio of partnerships and treat them the same way. Some advice:

• Don't invest more than you have to in order to maximize your potential gain.

• Watch partnerships carefully and be ready to end them as soon as it's clear they aren't going to reach their potential.
• Keep an eye out for potential partners and snag them early, while the cost of doing so is still low.

Your overall goal should be to create a portfolio that balances the value of intimate dealings with your partners against the cost of having to commit too many resources to capture that value. We've already seen how the Web can be a critical tool in achieving this goal. Look at how Intuit (INTU) has stitched together a rapidly changing set of financial service partners into what looks like a seamless supermarket at Quicken.com. The plumbing that connects these partners may look awful, but as long as the users aren't aware of it, you can transform the portfolio into a new product and service quickly and relatively cheaply.

Indeed, one way to think about the Internet Economy is not as hundreds of disjointed startups experimenting with new business models, but as a few very potent partnership portfolios. I mean, of course, the VCs (Kleiner Perkins), incubators (Idealab) and holding companies (CMGI) that lubricate the process with money but also pool the risk, help entrepreneurs leverage each other's expertise and experience, and connect the user communities of their holdings to multiply their network value.

Many of the big dot-com company mergers that have taken place this year were easier to accomplish because the parties all shared one or two major investors. Excite (ATHM) and AtHome, for example, were both part of the vast Kleiner Perkins keiretsu.

If you're in a traditional business, your new competitor is not some 25-year-old with $20 million in seed money. Your new competitor is all of Silicon Valley.

In creating your own portfolio of partnerships, you'll need to understand first what strengths and weaknesses you bring to the party, even before you test those of your potential partners. You may have a sterling brand name; impressive production, distribution, sales or marketing resources; or underutilized networks of loyal, educated, impassioned customers and suppliers.

Most likely, you will also have an encrusted industrial-economy bureaucracy finely tuned to react to potential partners as if they were deadly viruses entering the corporate bloodstream. That bureaucracy probably operates most effectively through legacy computing architectures and applications - precisely where you want it least.

Start by evaluating your current set of relationships. Then take a hard look at your ability to form new ones in a hurry. As Yahoo's Collet says: "A partner can't be someone who slows you down."

Larry Downes is an e-business consultant and coauthor of Unleashing the Killer App.